What Is a Training Needs Analysis?

By Brian Tr · Updated Oct 6, 2026

A training needs analysis (TNA) is the structured process a company uses to work out whether training is needed, who needs it, and what it should cover. It starts with a performance problem or a business goal, not with a course. The output is a short list of learning priorities that someone can fund.

If you sell learning products to companies, this matters more than it first appears. A TNA is often the document your buyer shows their boss to justify the purchase. Sellers who understand it can help build a stronger case. Sellers who ignore it get treated as noise. For the wider buying picture, see how companies buy training.

Key Facts

• The US Office of Personnel Management (OPM) describes a needs assessment as the process of identifying the gap between required and current performance, and then exploring why the gap exists and how to close it. It lists three levels: organizational, occupational and individual (OPM, Training Needs Assessment).
• OPM also states that sometimes training isn't the best solution, and that it's virtually never the only one. Other fixes include communicating expectations, supporting the work environment and checking job fit (OPM).
• The CIPD, the UK professional body for HR and L&D, separates a TNA (a one-off look at the need for a specific training activity) from an ongoing learning needs analysis, a health check on the organization's capabilities (CIPD, Learning needs analysis factsheet, updated August 2025).
• A published handbook chapter lays out a generic TNA as 16 steps in four phases: needs identification, needs specification, the TNA itself, and TNA evaluation (Surface, Training Needs Assessment, in The Handbook of Work Analysis, 2012).

What a training needs analysis actually is

Strip away the jargon and a TNA answers four plain questions:

1. What do we need people to be able to do?
2. What can they do now?
3. Why is there a gap, and is training the right fix?
4. If it is, who needs what, and in what order?

The terminology is messy. Training needs assessment, training needs analysis, gap analysis and front-end analysis all get used for roughly the same thing. A chapter in The Handbook of Work Analysis, written by Eric Surface, notes that people use these terms loosely and adopts "needs assessment" as the umbrella. It defines a TNA as a systematic process that links training requirements to deficiencies in individual, team or organizational performance, and uses them to set learning objectives (Surface, 2012).

Notice the order. The deficiency comes first. The learning objectives come after. A company that picks a course and then goes looking for a reason has skipped the analysis, which is how a lot of shelfware gets bought.

The three levels of analysis

The classic TNA looks at three levels: the organization, the task or job, and the person. The framing is usually traced to McGehee and Thayer's 1961 book Training in Business and Industry, and Irwin Goldstein's later work is cited alongside it. Surface's handbook chapter cites McGehee and Thayer (1961) and Goldstein (1993) when it discusses the person and organizational levels, so name both if you reference the model. Different bodies name the middle level differently: OPM calls it "occupational", most textbooks call it "task".

OPM's description of the levels is consistent with this. Organizational assessment evaluates performance and determines what skills, knowledge and abilities the agency needs. Occupational assessment examines the requirements of affected occupational groups. Individual assessment looks at how well a person is doing the job and which employees need what training (OPM).

Organizational analysis

This is the "should we, and can we?" level. Surface's chapter says a thorough organizational analysis links training objectives to organizational objectives, resources and constraints. It also checks whether the climate supports training. Is the business so busy that training will feel like a distraction? Are the resources there? Do existing policies, such as pay rules that reward some skills and not others, shape who signs up? (Surface, 2012)

For a seller, this is the level where deals quietly die. The program might be excellent, but if the company can't release people for four hours a week, nothing you offer will stick.

Task analysis

Here the question is what the job actually demands. The goal, in Surface's words, is to identify what must be learned so the capability transfers into performance. In practice that means breaking a role into tasks and the knowledge, skills and attitudes behind each one. Companies that already have a competency framework can skip much of this work. A continuous learning culture often keeps those frameworks current, which makes later analyses much faster.

Person analysis

Person analysis finds out who needs training on what. The same chapter notes that the "person" can also be a team or a department, and that this analysis comes in two forms: event-driven, triggered by a specific deficiency, and ongoing, where a systematic process keeps surfacing individual development needs (Surface, 2012). Annual performance reviews and development plans are the everyday version of the ongoing kind.

The process, step by step

There's no single official procedure. Surface notes there is no strong consensus on specific TNA procedures, though most processes share the same basic steps (Surface, 2012). Blending that chapter with OPM's guidance, a practical sequence looks like this:

1. Spot a trigger. Something prompts the question: a new regulation, a product launch, a reorganization, rising error rates, a failed earlier program, or a change in strategy. Surface's chapter lists fifteen kinds of trigger, from evaluation findings to new technology to attrition and customer complaints.
2. Check the need is real. Do a quick gap analysis. Is the performance gap actual or only perceived? What's the value of closing it, and the risk of ignoring it? Many requests end here, and that's a good outcome.
3. Ask whether training is the answer. OPM is blunt about this. Some gaps close through clearer expectations, a better work environment or better job fit. OPM adds that those same supports are needed if training is to produce lasting new behavior (OPM).
4. Plan the analysis. Agree the scope, the stakeholders and the success measures. OPM advises setting goals for the assessment and clarifying milestones before collecting data.
5. Collect and analyze data across the three levels (see the next section).
6. Specify the need and the options. Define the gap precisely: who, which skills, how big, how urgent. Compare options, including build, buy and non-training fixes. OPM adds a cost-benefit step so the team can build a business case.
7. Decide and act. Prioritize, set objectives, assign budget, choose solutions.
8. Evaluate. Check later whether the gap closed. Surface's fourth phase asks whether the issue that triggered the TNA was actually addressed.

Notice that vendor selection appears late, at the options stage. That's where you come in, and it explains why a vendor who shows up with a product before the need is defined has little to anchor on.

Where the data comes from

Good TNAs triangulate. The CIPD factsheet lists a wide range of methods, and says a mix used together gives better results than any one alone (CIPD):

• Existing organizational data, such as management information, learning platform records and performance management data
• Interviews and focus groups with stakeholders
• Informal conversations and attendance at team meetings
• Observation of real work
• Surveys of managers and employees
• Competency frameworks
• Documents such as business plans, job descriptions and work standards

The CIPD also warns that much of this data is sensitive, particularly when it exposes an individual's gaps, so confidentiality has to be respected.

The point for sellers is that your own product data can play a part. Usage and assessment data from a pilot is a legitimate input to a person-level analysis, if the customer chooses to use it that way. It's never a substitute for the analysis.

What a TNA produces

Done properly, the output isn't a binder nobody reads. The CIPD lists several outputs that can happen at the same time (CIPD):

• A report of overall learning needs for the organization or department
• A prioritized list of the performance gaps that matter most
• Learning and development plans once priorities and budgets are set
• Personal development plans aligned with available resources
• A list of potential solutions, including whether a formal intervention is needed at all

A prioritized gap plus a budget is, in effect, a buying brief. That's what makes the TNA so central to B2B learning sales.

TNA versus skills gap analysis

People use the two terms interchangeably. They shouldn't.

A skills gap analysis is one input to a TNA. The CIPD's capability questions describe it well: which capabilities will roles require, which exist in the workforce today, and what are the gaps between them (CIPD). The TNA takes that gap and asks the harder follow-ups. Why does it exist? Would training close it? Can the organization support the change?

The CIPD also treats an ongoing learning needs analysis as a continuous health check, and a TNA as a one-off look at a specific need. It helps to know which one your buyer means.

Why a TNA matters to an EdTech seller

A TNA is where the purchase gets justified. By the time a buyer asks for a quote, the real decisions about priority, audience and budget have often been made. Four consequences follow.

It changes who you're really competing with. The TNA's honest answer is sometimes "this isn't a training problem." Your competitor is then a process change, a manager conversation or doing nothing. See the corporate learning market for how much training budget never reaches outside vendors in the first place.

It sets the language of the deal. If the TNA describes a gap in "first-line manager coaching skills", your pitch should use those words. Matching their framing signals you've understood the problem and aren't reciting a catalog.

It defines success. A TNA ends with measurable performance gaps, which become the yardstick later. If you want a clean path to renewal, help the buyer pick a measure at this stage. How buyers then prove results is covered in proving training ROI.

It shapes the approval path. A documented need with a prioritized gap and a cost-benefit view gives the buyer something to carry through finance and procurement. See enterprise learning procurement for what happens next.

How to support a TNA honestly

There's a bad way to do this: handing buyers a "needs assessment" that happens to conclude they need your product. Experienced L&D managers spot it immediately. The L&D manager role exists partly to do this analysis independently, and they'll discount anything that looks rigged.

Here's what actually helps:

• Offer frameworks, not conclusions. A blank competency template, a sample interview guide or a checklist of data sources is useful whatever the buyer decides.
• Ask good diagnostic questions. What's the performance gap? How do you know? What's it costing? What have you tried? Who owns the result?
• Say when training isn't the answer. It feels like losing a sale. It builds trust, and OPM's own guidance says training is virtually never the only solution.
• Offer a scoped pilot with agreed measures. A short pilot against the gap the TNA found, with success criteria set in advance, lets the buyer test the analysis in the real world.
• Share data only with permission. Skill and assessment data about individuals is sensitive. Let the customer decide how it's used.

Different product types fit different levels. A cohort program usually answers a task-level need for a specific role. A self-paced library tends to serve ongoing person-level needs, which is part of the trade-off in cohort-based vs self-paced learning. A platform sits closer to the organizational level, since it serves delivery and tracking for everything else (LMS vs LXP).

When a full TNA doesn't happen

Plenty of purchases skip the formal process. Compliance training is the obvious case: the regulation defines the audience, the gap and the content. Surface's chapter makes this point, noting that when a regulation changes, the who and what are largely set and the analysis focuses on specifying content and delivery (Surface, 2012). Small companies rarely run a formal TNA either. The CIPD notes that identifying performance gaps still applies to smaller organizations, where people fill several roles, and suggests focusing on consulting business leaders and weighing costs and benefits (CIPD).

In those cases the TNA is informal. It happens in a founder's head or a one-page memo. The same questions still apply, so ask them even when nobody calls it a TNA. Timing matters too: a finished TNA usually feeds the next budget round, which is why L&D budget cycles decide when its recommendations can actually be bought.

Related reading

• How Companies Buy Training
• Proving Training ROI
• Enterprise Learning Procurement
• The Corporate Learning Market Explained